First, I have to wonder if the "desired" output you showed is "definitively correct" (in some sense), or whether it's simply "good enough to move along" (and other possible outputs - discarding a slightly different set of data rows - might be just as good, if not better).

Second, I'm curious about this distinctive property of your data sample: there is exact repetition of particular X values in the regions where certain data rows need to be discarded. Is this simply an artifact of how you created this data sample? If not, then would it happen to be a stable, reliable property of the actual data? If it is stable, this is something an algorithm could exploit. For example: 

(a) if the current x is lower than the previous x:
    (a.1) if the current x has been seen before, discard the current p
+oint
    (a.2) else discard the previous point.
(b) elsif the current x is identical to the prev. x (due to a.1 on pre
+v.iteration)
    (b.1) discard previous point

[download]
That logic will discard a different set of points than your "desired" output, but the selected points will be monotonic (if that property of repeating x values holds true).

If your pipeline streams get really huge, you might need to worry about not keeping track of too many "previously seen x values", but perhaps that's not a concern.

Here's the perl code for my approach (minus the input __DATA__): 

#! perl -slw
use strict;

printf "%s", scalar <DATA>;
my @points = map[ split ' ', $_ ], <DATA>;

my @modified;
my %xseen;

push @modified, shift @points;
$xseen{$modified[-1][0]} = undef;

while ( @points ) {
    if ( $points[0][0] < $modified[-1][0] ) {
        if ( exists( $xseen{$points[0][0]} )) {
            shift @points;
            next;
        }
        else {
            pop @modified;
        }
    }
    elsif ( $points[0][0] eq $modified[-1][0] ) {
        pop @modified;
    }
    push @modified, shift @points;
    $xseen{$modified[-1][0]} = undef;
}

print join ' ', @$_ for @modified;

[download]
And below is the output, pasted next to your "desired" output - mine is on the left side, and yours is on the right. Note that my code outputs 40 points compared to your 35 (I have the corresponding x's aligned so you can see the differences more easily). Also, there's one point where I kept the same x value as you, but selected a different y value at that x. 
0.0036759 0.4018006     0.0036759 0.4018006
0.0036962 0.4074616     0.0036962 0.4074616
0.0037064 0.4216399     0.0037064 0.4216399
0.0037166 0.4438854     0.0037166 0.4438854
0.0037268 0.4628417     0.0037268 0.4628417
0.0037370 0.4894152     0.0037370 0.4894152
0.0037471 0.4952320     0.0037471 0.4952320
0.0037675 0.4979326     0.0037675 0.4979326
0.0037879 0.5014988     0.0037879 0.5014988
0.0038082 0.5057747     0.0038082 0.5057747
0.0038184 0.5166984     0.0038184 0.5558402
0.0038286 0.5613627     0.0038286 0.5613627
0.0038388 0.6026338     0.0038388 0.6026338
0.0038490 0.6257104
0.0038693 0.6709805     0.0038693 0.6709805
0.0038795 0.6808655     0.0038795 0.6808655
0.0038897 0.6866130     0.0038897 0.6866130
0.0039202 0.6981425     0.0039202 0.6981425
0.0039406 0.7057251     0.0039406 0.7057251
0.0039508 0.7161121
0.0039610 0.7216518     0.0039610 0.7216518
0.0039712 0.7433434     0.0039712 0.7433434
0.0039813 0.7577987
0.0039915 0.7713018     0.0039915 0.7713018
0.0040017 0.7981869     0.0040017 0.7981869
0.0040119 0.8014242     0.0040119 0.8014242
0.0040221 0.8264223     0.0040221 0.8264223
0.0040526 0.8290191     0.0040526 0.8290191
0.0040628 0.8323083     0.0040628 0.8323083
0.0040933 0.8361688     0.0040933 0.8361688
0.0041035 0.8409814     0.0041035 0.8409814
0.0041137 0.8527880
0.0041239 0.8591068     0.0041239 0.8591068
0.0041341 0.864785 
0.0041443 0.8760895     0.0041443 0.8760895

[download]
(As a general rule, other things being relatively equal, I'd expect it might be desirable to discard fewer data points - but I suppose that depends on the nature of your test equipment.)

(updated to fix the wording in the 2nd paragraph and pseudo-code)

In reply to Re: How would you code this? by graff
in thread How would you code this? by BrowserUk

Title:
Use:  <p> text here (a paragraph) </p>
and:  <code> code here </code>
to format your post, it's "PerlMonks-approved HTML":


  • Posts are HTML formatted. Put <p> </p> tags around your paragraphs. Put <code> </code> tags around your code and data!
  • Titles consisting of a single word are discouraged, and in most cases are disallowed outright.
  • Read Where should I post X? if you're not absolutely sure you're posting in the right place.
  • Please read these before you post! —
  • Posts may use any of the Perl Monks Approved HTML tags:
    a, abbr, b, big, blockquote, br, caption, center, col, colgroup, dd, del, details, div, dl, dt, em, font, h1, h2, h3, h4, h5, h6, hr, i, ins, li, ol, p, pre, readmore, small, span, spoiler, strike, strong, sub, summary, sup, table, tbody, td, tfoot, th, thead, tr, tt, u, ul, wbr
  • You may need to use entities for some characters, as follows. (Exception: Within code tags, you can put the characters literally.)
            For:     Use:
    & &amp;
    < &lt;
    > &gt;
    [ &#91;
    ] &#93;
  • Link using PerlMonks shortcuts! What shortcuts can I use for linking?
  • See Writeup Formatting Tips and other pages linked from there for more info.